In 3GPP Long Term Evolution (LTE), consideration is given to maintaining orthogonality between mobile stations (UE: user equipment) by allocating an orthogonal radio resource to each mobile station (UE) when transmitting and receiving data (a radio resource is an area uniquely defined by time and frequency; radio resources are set by dividing time and frequency into discrete areas for allocation to different mobile stations, so that one resource will not overlap between two mobile stations).
During transmission/reception of an uplink signal, in order to eliminate interference between mobile stations (UE) within the cell of a base station (Node B) so that the uplink signals can be demodulated correctly by the base station (Node B), it is essential that the base station's receive timing of an uplink signals from each of a plurality of mobile stations (UE) fall within a guard interval called a “cyclic prefix (CP).” At the same time, regardless of whether synchronization in data transmission/reception is actually being maintained, synchronization is assumed to be guaranteed if a receive timing falls within a predetermined timer period (i.e. during a timer is running). Based on this assumption, a state in which a receive timing falls within the timer period is judged to be a deemed in-sync state (i.e. a mobile station is assumed to be uplink synchronized), and a state in which a receive timing does not fall within the timer period (i.e. a timer expires) is judged to be a deemed out-of-sync state (i.e. a mobile station is assumed to be NOT uplink synchronized).
A mobile station (UE) which has been determined to be in a deemed out-of-sync state sends a Non-sync RACH (Non-synchronized Random Access Channel), in which a plurality of mobile stations (UE) compete for and use common radio resources, before transmitting an uplink signal. The mobile station then receives from the base station a timing advance (TA) for adjusting its transmit timing. According to the TA, the mobile station adjusts its transmit timing and finally synchronizes the uplink signal (i.e. the uplink signal is received within a CP at the base station).
Since each mobile station (UE) must secure synchronization while it is transmitting an uplink signal, a timing advance (TA) is notified from the base station to each mobile station, either at constant intervals or triggered by the occurrence of a specific event (e.g., a rapid change in the traveling speed of the mobile station). The reference time period from when the timing advance (TA) is last updated until the mobile station is judged to have returned to an out-of-sync state is either notified from the base station in the system information as the cell specific value or is pre-defined as a fixed value. The reference time is monitored at the base station and each of the mobile stations (UE) through use of a timer. Upon a timeout of the timer (i.e., when the reference time period described above expires), the mobile station (UE) is judged to have transited from a deemed in-sync state to a deemed out-of-sync state.
For the purpose of judging whether a mobile station (UE) is in a deemed in-sync state or a deemed out-of-sync state, the base station controls as many timers as the number of mobile stations (UE) under its management. The mth timer held by the base station (where m is an integer between 1 and M, and M is a natural number indicating the number of mobile stations (UE) managed by the base station) corresponds to the timer held by the mth mobile station (UE).
The plurality of timers controlled by the base station are set to the same length of time, so are the timers held by the plurality of mobile stations. The base station timers and the mobile station timers are set to a timer length such that synchronization can be guaranteed by the mobile station (UE) that is traveling at the highest speed (e.g., 350 km/h) of all the mobile stations (UE) supported by the base station. The timer length is therefore shorter than the minimum length of time over which this mobile station (UE) will become out-of-sync. A determination between a deemed in-sync state and a deemed out-of-sync state is made solely relying on the state of the timer, regardless of the actual traveling speed of the mobile station (UE). Non-patent Literature 1 discloses an example of a process for adjusting a transmit timing during transfer of an uplink signal in the 3GPP Long Term Evolution (LTE) described above.
When data is generated for transmission to the base station, a mobile station (UE) in a deemed in-sync state first transmits a Scheduling Request (SR) to the base station to request a radio resource over which to transmit the data, using a radio resource specific to the mobile station (UE). One method that can be used to assign a radio resource specific to a mobile station (UE) is to periodically assign a radio resource over which to transmit an SR to each of the mobile stations (UE) that are in a deemed in-sync state. On the other hand, when a mobile station (UE) in a deemed out-of-sync state transmits an SR, it first transmits a Non-sync RACH and receives a timing advance (TA) for controlling the transmit timing and, at the same time, assigns a radio resource specific to the mobile station (UE) over which to transmit the SR.
It is clear from the foregoing that a mobile station (UE) in a deemed out-of-sync state suffers a longer latency (or a delay due to waiting time) before it can initiate data transmission than a mobile station in a deemed in-sync state. This is because the former mobile station additionally requires a step of transmitting a Non-sync RACH before being able to perform a step of transmitting an SR. In addition, since orthogonality between mobile stations (UE) is not guaranteed for a Non-sync RACH, a collision may occur between mobile stations (UE). If a collision occurs, the transmitted Non-sync RACH may not be detected by the base station, in which case the mobile station (UE) must retransmit a Non-sync RACH. This further increases the latency.    Non-patent Literature 1 3GPP RAN WG2 Contribution [R2-063401.doc NTT DoCoMo]http://www.3gpp.org/ftp/tsg_ran/WG2_RL2/TSGR2—56/Documents/    Non-patent Literature 2 3GPP RAN WG1 Contribution [R1-063377.doc Nokia]http://www.3gpp.org/ftp/tsg_ran/WG1_RL1/TSGR1—47/Docs/    Non-patent Literature 3 3GPP RAN WG1 Contribution [R1-063405.doc Siemens]http://www.3gpp.org/ftp/tsg_ran/WG1_RL1/TSGR1—47/Docs/
In the related art described above, the timer length is set based on the time in which synchronization is guaranteed by a rapidly traveling (for example, at a speed of 350 km/h) mobile station (UE). This leads to a problem in which a mobile station (UE) that is standing still or traveling at a low speed may be judged to be in a deemed out-of-sync state upon a timeout of the timer, even though the actual uplink synchronization is being maintained.
Furthermore, if another data occurs at the mobile station (UE) and an uplink signal must be transmitted during a period in which the timer has timed out but synchronization is actually being maintained, the latency before the data can be transmitted becomes even longer.
This is because a static or slow-moving mobile station (UE) is judged to be out-of-sync based on the timer, even though it is actually in sync and can transmit a Scheduling Request (SR) using a radio resource specifically assigned to it. In such situation, the mobile station needs first to transmit a Non-sync RACH to receive a timing advance (TA) from the base station (Node B), so that it can be assigned a radio resource over which to transmit the SR according to the timing advance (TA).